Paper Info
Title
High-quantile modeling for customer wallet estimation and other applications
Abstract
In this paper we discuss the important practical problem of customer wallet estimation, i.e., estimation of potential spending by customers(rather than their expected spending). For this purpose we utilize quantile modeling, whose goal is to estimate a quantile of the discriminative conditional distribution of the response, rather than the mean, which is the implicit goal of most standard regression approaches. We argue that a notion of wallet can be captured through high quantile modeling (e.g, estimating the 90th percentile), and describe a wallet estimation implementation within IBM's Market Alignment Program (MAP). We also discuss the wide range of domains where high-quantile modeling can be practically important: estimating opportunities in sales and marketing domains, defining 'surprising' patterns for outlier and fraud detection and more. We survey some existing approaches for quantile modeling, and propose adaptations of nearest-neighbor and regression-tree approaches to quantile modeling. We demonstrate the various models' performance in high quantile estimation in several domains, including our motivating problem of estimating the 'realistic' IT wallets of IBM customers.
Year
DOI
Venue
2007
10.1145/1281192.1281297
KDD
Keywords
Field
DocType
it wallet,customer wallet estimation,high-quantile modeling,wallet estimation implementation,expected spending,ibm customer,implicit goal,quantile modeling,high quantile modeling,high quantile estimation,nearest neighbor,data mining,regression tree
Econometrics,Data mining,IBM,Conditional probability distribution,Regression,Computer science,Outlier,Quantile,Artificial intelligence,Discriminative model,Percentile,Machine learning
Conference
Citations 
PageRank 
References 
12
1.42
8
Authors
4
Name
Order
Citations
PageRank
Claudia Perlich152345.01
Saharon Rosset21087105.33
Richard D. Lawrence332221.24
Bianca Zadrozny41585135.66